Comparative evaluation of predictive value of three caries activity tests-snyder, lactobacillus count and cariostat in mixed dentition children with and without caries.

INTRODUCTION: Dental caries is a disease with multifactorial etiology. Early detection and preventive measures are essential for caries control. Caries activity tests are one of the methods of early detection, but all these are based on microbiological study. AIMS: Compare the sensitivity and specificity of Snyder test (ST), lactobacillus count (LBC) and Cariostat in mixed dentition children with and without caries.
MATERIALS AND METHODS: 90 healthy mixed dentition children (79 boys and 11 girls) were selected and were divided into two groups, one with Decayed missing filled teeth (DMFT) 0 and another group with DMFT 3-5. The salivary samples were collected for ST and LBC and plaque samples were collected for Cariostat. Caries activity was assessed.
RESULTS: The data obtained were analyzed for sensitivity, specificity, positive predictive value, negative predictive value and over all accuracy.
CONCLUSIONS: Our present study showed that the LBC was found to have a higher correlation in caries as the caries increased the count also increased with positive caries activity. A good correlation was found between Snyder and LBC for the caries activity 1, 2 and 3. But no correlation was seen for score 0. To conclude, a combination of all these tests was found to be good for assessing caries risk groups.

It is well known that dental caries is a disease with a multifactorial etiology in which it begins with the loss of ions from the apatite crystals and ends in cavitation.

Carious lesions can be easily detected by a clinical examination with a mouth mirror and explorer coupled with bitewing X rays. However, a clinical examination neither predicts caries activity nor indicates a patient′s susceptibility to dental caries. All the caries activity tests developed so far is based on microbiological aspect of dental caries. The need for a caries susceptibility test is to run a caries preventive program and without using microbiological methods, it is like running a weight control program without a scale.[1]

Nearly 10-15% of children worldwide within school dental service present a high caries activity, but they take up a large population of the resources available for dental care.[2] So it is important to identify these children before they develop a large number of carious lesions and provide adequate preventive measures to stop the destruction of their teeth.

Caries activity tests do not measure the innate susceptibility of a person to caries; currently they only measure the degree of the local environment challenge towards the probability of new carious lesions.[1]

Hadley[3] developed a method of selective growth of lactobacillus (LB), Snyder[4] developed a simple colorimetric test to assess the relative number of LB in saliva. Cariostat[56] is a simple colorimetric test developed to assess the acid produced by bacteria in plaque. All these tests concentrate on estimating streptococcus mutans and lactobacilli in saliva and plaque.

The aim of the study is to compare the sensitivity and specificity, positive predictive value (PPV) and negative predictive value (NPV) and overall accuracy (OA) of Snyder test (ST), lactobacillus count (LBC) and Cariostat towards caries activity in mixed dentition children with and without caries.

Materials and Methods

A total 90 healthy mixed dentition children (79 boys and 11 Girls) between the age group of 8-12 years were selected from the school and written informed consent of the parents as well as the school authorities were taken before including in the study.

Children with history of antibiotic consumption in the recent past and any additional fluoride use, undergoing orthodontic treatment or wearing any appliance were excluded from the study.[7] The children were brought in batches and clinically examined for dental caries using a good plain mouth mirror and a sharp explorer and were divided into two groups out of which 45 were without caries and 45 with decayed missing filled teeth (DMFT) 3-5. Dental caries was recorded as per criteria described by Klein.[8] In the first visit children were trained to chew elastics on both sides for the collection of saliva. The salivary samples were collected between 10 Am and 11 Am to give sufficient time after breakfast to prevent any alteration in physiochemical properties of saliva. All the children were subjected to following tests:

ST

Cariostat

LBC

ST

Bromocresol green dextrose (BCG) agar (Hi-Media Pvt Ltd) media was used for ST which had bromocresol green as an indicator. The subjects were asked to eliminate any existing saliva in the oral cavity by swallowing. Stimulated saliva was collected in a calibrated plastic container. The collected saliva was shaken and 0.2 ml of saliva was pippeted into the agar media and mixed. The petridishes were incubated at 37°C and colour change was observed at intervals of 24, 48 and 72 h after incubation against a white background. The change in colour was classified into 3 categories which denoted the corresponding caries activity [Table 1].[19]

Table 1

Caries activity scores for ST

Cariostat

Each 2 ml test liquid consists of:

20% Sucrose as carbohydrate

2% Tryptose

0.5% Sodium chloride

0.02% NaN3

Indicators-Bromocresol green and purple.

Method and collection of plaque samples.[5610]

In each subject, collection of plaque was done before the collection of saliva. Plaque samples were collected from the buccal surfaces of the upper first molars with a sterile cotton bud. One half of the plaque samples collected from one half of the maxillary teeth was placed in bormocresol green indicator and the other half of the plaque sample in bormocresol purple and incubated for 48 h at 37°C. Scoring was done by comparing the test tubes against a white background for colour change in pH [Table 2].[5610]

Table 2

Caries activity scores for cariostat

LBC

Rogosa SL agar (Hi-Media Pvt Ltd) was the media used. The Wooden spatula method described by Kohler and Bratthal[11] was used by placing a sterile wooden spatula placed in the floor of the mouth and it was made sure that both sides of the spatula were moistened with saliva and pressed against the surface of the Rogosa agar. Inoculated plates were incubated in an anaerobic jar producing microaerophilic atmosphere of 85% nitrogen, 5-10% carbon dioxide and 5-10% oxygen for 48 h at 37°C. The colonies formed on Rogosa SL agar were identified as a medium to large in size, milky white, round, convex with a smooth border and surface. Colony forming units (CFU), with a magnifying glass, in a circular area of 1.5-2 cm of spatula impression formed by considering a distance of 1 cm from the tip of rounded end.

The CFU were classified as microbial count per ml of saliva [Table 3].

Table 3

Lactobacillus CFU were classified as microbial count per ml of saliva

Results

Statistical analysis

The data obtained were analysed for the following:

Sensitivity

Specificity

PPV

NPV

OA

Out of 45 subjects examined in caries free group (dmft zero) in cariostat, caries inactive (score 0) of pH 6.8 was shown in 37 subjects (82.2%), caries active (score 1) of pH 5.2 in 3 subjects (6.7%), pH 4.5 in 2 subjects (4.4%) (score 2), pH 3.6 in 3 subjects (6.7%) (score 3). In 45 caries group children, the result was 30 subjects (66.7%) (score 0), caries active score 5.2 pH in 2 subjects (4.4%) (score 1), 4.5 pH in 5 subjects (11.1%) (score 2) and 3.6 pH in 8 subjects (17.8%) (score 3) [Table 4].

Table 4

Frequency distribution of subjects in cariostat with caries activity

Out of 45 caries free subjects caries inactivity was shown in 37 subjects (82.2%) (score 0), limited caries activity in 4 subjects (8.9%) (score 1), definitive in 4 subjects (8.9%) (score 2) and none showed marked caries activity (score3). Out of 45 subjects in caries group, caries inactive score was seen in 33 subjects (73.3%) (score 0), limited caries activity in 2 subjects (4.4%) (score 1), definitive in 8 subjects (17.8%) (score 2) and marked in 2 subjects (4.4%) (score 3) [Table 5].

Table 5

Frequency distribution of subjects in snyder test with caries activity

Out of 45 caries free subjects, caries inactive count 0 was seen in 8 subjects (17.8%), caries active count 1-20 was seen in 28 subjects (62.2%) (score 1), count 21-40 was seen in 9 subjects (20.0%) (score 2), and none of the subjects showed counts 41-100 (score 3).

Out of 45 subjects in caries group, caries inactive count 0 was seen only in 1 subject (2.2%) (score 0), caries active counts 1-20 were seen in 20 subjects (44.5%) (score 1), count 21-40 was seen in 19 subjects (42.2%) (score 2), counts 41-100 was seen in 5 subjects (11.1%) (score 3) [Table 6].

Table 6

Frequency distribution of subjects in lactobacillus count with caries activity

Cariostat versus ST with dmft/DMFT 0

In caries free group, out of 8 subjects, which were shown to be caries active (scores 1-3) by Synder, only 1 subject of Synder correlated with 8 subjects of cariostat in caries active group. Sensitivity 1/8 = 12.5%, PPV 1/8 = 12.5%. Out of 37 subjects, which were shown to be caries inactive by Synder, 30 subjects of synder correlates with that of 37 subjects of Cariostat in caries inactive group. Specificity 30/37 = 81.1%, NPV 30/37 = 81.1%. The OA of this validity test showed 31/45 = 68.9% [Table 7].

Table 7

Predictive value of cariostat versus snyder with dmft/dmft=0

Cariostat versus ST with dmft/DMFT 3-5

In caries group dmft/DMFT,[3-5] out of 12 subjects, which were shown to be caries active (scores 1-3) by Synder, only 6 subjects of Synder correlated with 15 subjects of Cariostat in caries active group. Sensitivity 6/15 = 40%, PPV 6/12 = 50%. out of 33 subjects which were shown to be caries inactive by Synder, 24 subjects of Synder correlates with that of 30 subjects of Cariostat in caries inactive group, Specificity 24/30 = 80%, NPV 24/33 = 72.7%. The OA of this validity test showed 30/45 = 66.7% [Table 8].

Table 8

Predictive value of cariostat versus snyder with dmft/dmft=3-5

Predictive value of snyder versus LBC with dmft/DMFT 0

In caries free group, out of 37 subjects, which were shown to be caries active (scores 1-3) by LBC, 7 subjects of LBC correlated with that of 8 subjects of Synder in caries active group, sensitivity 7/8 = 87.5%, PPV 7/37 = 18.9%. Out of 8 subjects, which were shown to be caries inactive by LBC, only 7 subjects of LBC correlated with that of 37 subjects of Synder which were shown to be caries inactive, Specificity 7/37 = 18.9%, NPV 7/8 = 87.5%. The OA of this validity test showed 14/45 = 31.1% [Table 9].

Table 9

Predictive value of snyder versus lactobacillus count with dmft/dmft=0

Predictive value of snyder versus LBC with dmft/DMFT 3-5

In caries group dmft/DMFT[3-5] out of 44 subjects, which were shown to be caries active (scores 1-3) both Synder and LBC showed good correlation in all 12 cases, sensitivity 12/12 = 100%, PPV 12/44 = 37.3%. Only 1 subject showed caries inactivity in LBC, when compared to 33 subjects in Synder. Specificity 1/33 = 3%, NPV 1/1 = 100%. The OA of this validity test showed 13/45 = 28.9% [Table 10].

Table 10

Predictive value of synder versus lactobacillus count with dmft/dmft=(3-5)

Predictive value of cariostat versus LBC with dmft/DMFT 0

In caries free group, out of 37 subjects, which were shown to be caries active (scores 1-3) by LBC, 7 subjects of LBC correlated with that of 8 subjects of cariostat in caries active group. Sensitivity 7/8 = 87.5%, PPV 7/37 = 18.9%. Out of 8 subjects which were shown to be caries inactive by LBC, only 7 subjects of LBC correlated with that of 37 subjects in cariostat. Specificity 7/37 = 18.9%, NPV 7/8 = 87.5%. The OA of this validity test showed 14/45 = 31.1% [Table 11].

Table 11

Predictive value of Cariostat versus lactobacillus count with dmft/dmft=0

Predictive value of cariostat versus LBC with dmft/DMFT 3-5

In caries group dmft/DMFT,[345] out of 44 subjects, which were shown to be caries active (scores 1-3) by LBC, both cariostat and LBC showed good correlation in all 15 cases. Sensitivity 15/15 = 100%. PPV 15/45 = 34.1%. Only 1 subject showed caries inactivity in LBC when compared to 30 subjects in cariostat. Specificity 1/30 = 3.3%, NPV 1/1 = 100%. The OA of this validity test showed 16/45 = 35.5% [Table 12].

Table 12

Predictive value of cariostat versus lactobacillus count with dmft/dmft=(3-5)

From the above results, both Cariostat and Synder showed a high correlation when it was caries inactive (score was 0). But Cariostat showed higher caries activity than Synder when caries active (scores 1-3). Both Cariostat and Synder had higher correlation with LBC when the score was 1-3. A low correlation was seen with LBC by both Cariostat and Synder when score was 0.

Discussion

The caries activity refers to the increment of active lesions, including new and recurrent lesions that occur over a stated period of time. Susceptibility refers to the inherent propensity of the host and the target tissue, the tooth to be affected by caries processes. Dental caries is established in the mouth long before it becomes clinically manifested as a viable lesion, which means that it should be possible to assess the seriousness of the factors related with caries before a viable lesion can develop. Identifying and measuring the magnitude of these factors is the primary aim of the caries activity test. These tests can be utilised in individuals who are at high risk of developing dental caries, which can help in focusing caries preventive measures on them.[1] This study was done to compare the efficacy in predicting caries in the same subjects of both caries free and carious group with dmft 3-5.

It was first described by Snyder based on acid production where saliva is inoculated into agar medium of pH 4.7-5 with bromocresol green indicator which changes from green to yellow at lower pH. This estimates the number of both aciduric and acidogenic organisms in saliva. BCG dextrose agar was preferred to Snyder′s media as it is prepared from purified pancreatic digest of casein and the easy availability and handling.[12] Newbrun reported that ST is based on aciduric microorganisms, a quantitative colorimetric test used for group correlation, but unsatisfactory for individuals. This can be used for diet councelling and for monitoring the effectiveness of preventive procedures.[9]

The name was developed from the summary of the procedures and test liquid used and it determines the state of caries.[5] The test liquid contains two indicators bromocresol green and bromocresol purple, which were added separately as it indicates two different levels of pH ranges. The plaque samples were collected before saliva to prevent plaque dislodgement from the buccal surface of upper molar as it was the best location to reflect the entire oral cavity.[5]

The media used was Rogosa SL agar and this was preferred over tomato juice agar as the latter inhibited the growth of the other organisms, in their media. The wooden spatula method was preferred over platinum loop inoculation due to easy availability and little microbiological expertise.[11] This method has high validity in children at risk of dental caries, when compared with both sides of spatula and different sampling times.[13]

In this study when caries activity was assessed from the same subjects by 3 different methods, it was found that caries activity in 37/45 subjects (scores 1-3) in LBC, 8/45 in ST and cariostat in caries free group. In caries active group, LBC showed 44/45, Snyder showed 12/45 and cariostat 15/45 subjects. Increased number of LBC was attributed to the fact that they are the important pathogens in the progression of caries with Strep. mutans as the initiator of caries.[14] Cariostat and Snyder couldn′t show that many cases because these tests are depended on acid produced by cariogenic organisms, which produce colour change in the media. In this study one subject of dmft 3-5 and 8 subjects with DMFT 3-5 showed absence of lactobacilli. This could be attributed to the fact that LB have a difficulty in attaching themselves and they need mechanical retentive areas where aciduric environment can be established.[15] Absence of LB is seen in caries free group as reported by Hill and Blayney.[16]

Massler[17] reported that caries incidence peaks at ages 4-8 and 11-19 with a marked decrease between 8 and 11 due to changes in oral environment, eruption, exfoliation, number of teeth present and food habits. A similar observation was seen in the study with low level of LBC. A good correlation was seen between ST and LBC with dmft 3-5 than dmft 0. When the dmft was 0, the sensitivity was 87.5%, Specificity was 18.9%, PPV was 18.9%, NPV was 87.5% and OA was 31.1%. In dmft group 3-5, the sensitivity was 100%, Specificity 3%, PPV of 27.3%, NPV of 100% and over all accuracy of 28.9%.

A good correlation was seen between Cariostat and Snyder when it was caries inactive, but there was no good correlation with dmft 3-5 in caries active group as reported by Shimono.[5] In dmft 0, the sensitivity was 12.55, Specificity was 81.1%, PPV 12.5%, NPV 81.1% and OA of 68.9%, but in the dmft 3-5 group, sensitivity was 40%, Specificity was 80%, PPV 50%, NPV 72.7% and OA 66.7%. When Cariostat was compared to LBC in the caries free group, sensitivity was 87%, Specificity 18.9%, PPV18.9%, NPV 87.5% and OA 31.1%. In the dmft 3-5 group, the sensitivity was 100%, Specificity 3.3%, PPV 34.1%, NPV 100% and OA of 35.5%.

None of the caries activity tests are highly reliable as indicators of expected caries increments as it measures single parameters such as acid produced or colony counting of bacterial species and as dental caries is a multifactorial disease and caries predictive tests do not encompass those factors like fluoride exposure, maturation of enamel or immune protection.[1]

Sizhen Shi et al.[18] reported that caries activity detection plays a vital role in preventing and treating caries in preschool age children. Results from the Dentocult SM application show that the method can detect the presence of caries activity and forecast its progress, and is correlated to caries severity with high statistical significance. Dentocult LB is less desirable, and Dentobuff Strip the least.

Bhasin et al.[19] studied Oratest, a simple chair side caries activity test and reported that higher level of infection, lesser was the time taken for the colour change of the expectorated milk reflecting higher oral microbial levels. Significant overall correlations were observed in oratest data and deft/DMFT or Plaque index. Oratest facilitates the clinical management of patients as they determine the need and extent of personalised preventive measures. It helps in identifying high risk groups and helps to motivate and monitor the effectiveness of educational programmes relating to dietary and oral hygiene procedures.

Kaur et al.[20] studied non microbial salivary caries activity parameters and salivary biochemical indicators and reported that normal level of hydration, higher values for flow rate, pH, buffering capacity of saliva lead to good oral health and reduced caries occurrence. Increased salivary viscosity increases caries incidence. Salivary biochemical indicators like calcium, phosphorus, alkaline phosphatase also play their respective role in determining caries susceptibility of an individual. These can be used as salivary parameters for caries risk assessment.

These tests have a great value in monitoring the effectiveness of patient education in dietary and plaque control measures and restorative procedures, with subsequent caries activity test results help as a useful index for the success of treatment and special prevention attention. From this study it shows that LBC shows good predictive capacity in both caries free and caries active group when compared to Snyder and cariostat. To assess the increment of caries in these subjects, a longitudinal study should be done to know the predictive value of caries activity. A combination of these tests will prove to be very useful for assessing caries activity in children.

Conclusions

When ST was compared with Cariostat, sensitivity was found to be 12.5% and specifiticity was 81.1% in the dmft 0 group and sensitivity was 40% and specificity was 80% in the dmft 3-5 group.

When LBC was compared with ST, sensitivity was 100% and specificity was 3% in the dmft 3-5 group and sensitivity and specificity were 87.55% and 18.9% in the dmft 0 group.

When LBC was compared with Cariostat, sensitivity was 100% and specificity was 3.3% in dmft 3-5.

LBC was found to have higher correlation with caries, the count increased with the increase in caries and showed positive caries activity.

A good correlation was seen between Snyder and LB count with caries active score 1-3 but no correlation when caries inactive score 0.

A combination of all these tests were found to be good for assessing caries risk groups and planning preventive measures and future assessment.